Contributions to Deep Learning Models - RiuNet
... Graphical depiction of the Local-DNN model. Several patches are extracted from the input image and they are fed into a DNN which learns
a probability distribution over the labels in the output layer. The final
label of the image is assigned using a fusion method that takes into
account all the patch ...
A tale of two stories: astrocyte regulation of
... Mechanisms of short-term presynaptic plasticity
Despite its apparent simplicity, the Tsodyks-Markram (TM) model (equations 1-2) can generate
surprisingly complex synaptic dynamics including multiple mechanisms of short-term plasticity
among which are facilitation and depression. Nonetheless, the occ ...
... normalised by the number of outputs only, the average over all rows (all i) returns the same as
grad(input, output). This function is useful for implementing on-line teaching algorithms.
mlp_gradij computes gradients of network outputs, i.e the derivatives of outputs w.r.t. active
weights, at given ...
Mechanisms of Leptin Action and Leptin Resistance
... the SH2 proteins that they recruit. There are
three conserved residues on the intracellular
domain of LRb: Tyr985 , Tyr1077 , and Tyr1138 .
Data from our and other labs suggest that all
three of these sites are phosphorylated and
contribute to downstream leptin signaling (8,
54, 55, 60, 60a).
Dowe2010_MML_Handboo.. - Clayton School
... Furthermore, defining a prefix code to be a set of (k-ary) strings (of arity k, i.e.,
where the available alphabet from which each symbol in the string can be selected
is of size k) such that no string is the prefix of any other, then we note that the 2n
binary strings of length n form a prefix code ...
as a PDF
... Some other approaches include genetic fuzzy neural
networks and genetic fuzzy clustering, among others
Calcium Transients in the Garter Snake Vomeronasal Organ
... patterns of activity displayed a scattered appearance with a
heterogeneous organization in which it was possible to find
nonuniform foci of activity distributed in multiple epithelial
regions separated by silent sectors. Within each lamina there
were important variations in the amplitude and time co ...
Understanding the process of multisensory integration
... than those from cues that are temporally displaced from one another. However, the
present results from studies of cat SC neurons show that this "temporal principle" of
multisensory integration is more nuanced than previously thought and reveal that
the integration of temporally-displaced sensory res ...
neuronal reward and decision signals: from theories to data
... implemented in the brain in various neuronal reward signals, and thus does seem to have a physical basis. Although
sophisticated forms of reward and decision processes are far
more fascinating than arcane fundamental variables, their
investigation may be crucial for understanding reward processing. ...
Towards Smart User Models for Open Environments
... context, personalised and adaptive human-system interfaces have become a key
requirement in understanding user requirements [Luck, et al.; 2003a], [Murray; 2002].
As G. Fischer says: The challenge in an information-rich world is not only to make
information available to people at any time, at any pl ...
PART 1 - FTP Directory Listing
... To create the skeleton of CRONOS, the human skeleton was copied as accurately as
possible at life size.3 The bones were constructed from a new type of thermoplastic known in the
UK as Polymorph and in the US as Friendly Plastic, which softens and fuses at 60 degrees and
can be freely hand moulded un ...
THE ELECTRODE-TISSUE INTERFACE DURING RECORDING
... requirements over the years. Dr. Landau thanks for helping spark my interest in
electrochemisty, even though I am not good at it. I really enjoyed your class. Dawn
thanks for your willingness to help and for sharing your wealth of knowledge in the area
of microelectrode recording. Dr. Durand thanks ...
Outputs of Radula Mechanoafferent Neurons in Aplysia are
... The transmission of sensory information from the periphery
to the nervous system is modulated both at the level of primary
sensory afferents (Brooke et al. 1997; Gu and MacDermott
1997; Hill et al. 1997; Passaglia et al. 1998; Pasztor and
Macmillan 1990) and at various stages of processing in the
Feeding Stimulants Activate an Identified Dopaminergic Interneuron
... connections with the rest of the CNS (but see Kabotyanski et
al. 1994). Second, dopaminergic neurons, whose activation
mimics the effects of dopamine superfusion, have rarely
been identified. Third, there is diversity of dopamine receptors both within and across species (Ascher 1972; Berry
and Cottr ...
... A new post processing method is developed in  for maintaining a good interpretability-accuracy
trade-off in linguistic fuzzy systems, which performs rule selection and membership function tuning by
focusing on the Pareto zone having most accurate solutions but the least number of possible rules. ...
Financial Time Series Forecasting Using Improved Wavelet Neural
... outperforms each of the two individual models.
Traditional learning approaches have difficulty with noisy, non-stationary time
series prediction. For this reason,  uses a hybrid model which combines
symbolic processing and recurrent neural networks to solve the problem. The
model converts the ti ...
Neural modeling fields
Neural modeling field (NMF) is a mathematical framework for machine learning which combines ideas from neural networks, fuzzy logic, and model based recognition. It has also been referred to as modeling fields, modeling fields theory (MFT), Maximum likelihood artificial neural networks (MLANS).This framework has been developed by Leonid Perlovsky at the AFRL. NMF is interpreted as a mathematical description of mind’s mechanisms, including concepts, emotions, instincts, imagination, thinking, and understanding. NMF is a multi-level, hetero-hierarchical system. At each level in NMF there are concept-models encapsulating the knowledge; they generate so-called top-down signals, interacting with input, bottom-up signals. These interactions are governed by dynamic equations, which drive concept-model learning, adaptation, and formation of new concept-models for better correspondence to the input, bottom-up signals.